STINSON, RAHMOELLER, TECKLENBERG, COLVIN, JONES AND PIERCE 
431 
FR 1300). Animals which balked during the 
study or pulled against the leash were excluded 
from analysis. 
The pressure transducers were calibrated be- 
fore and after implantation in vitro at 38° C. 
Static calibration against a mercury column 
was found to be constant and linear from zero 
to 300 mm Hg ( ± 1 % ) without hysteresis, and 
the frequency response of the pressure record- 
ing system was linear to 80 cps ( ± 5 % ) . Zero 
stability, though constant with some gauges, 
was not found to be reliable; thus, changes 
in left ventricular end-diastolic pressure 
(LVEDP) are reported as deviations from a 
preset baseline arbitrarily set to 4-5 mm Hg 
(catheterization studies in several dogs showed 
LVEDP to range from 2 to 8 mm Hg). Since 
zero baselines in vitro were found to be stable 
over a minimum period of two hours in the lab- 
oratory, delta values for LVEDP during the 10 
minute periods of study were considered to be 
valid. The rate of change of left ventricular 
pressure (LV dp/dt) was obtained with an ac- 
tive differentiator having a linear frequency re- 
sponse and constant phase error of two msec to 
80 cps. Pressure and other measurements were 
averaged over several cardiac cycles selected 
during expiratory pauses. 
Aortic flow was measured with a Biotronex 
Model 610 Pulsed-Logic flowmeter whose static 
calibration was linear to 20 L/min and whose 
frequency response was linear to 50 cps (± 
8%). Flow transducers were calibrated pre- 
and postimplantation, and in some instances in 
situ after sacrifice of the animal. Late diastolic 
flow was taken to represent zero flow, and 
stroke volume was obtained by electronic inte- 
gration. 
For the determination of left ventricular di- 
ameter the transmitting crystal was pulsed for 
one microsecond at 5.6 KHz. The received sig- 
nal from the opposite crystal was rectified, am- 
plified, and detected by a tracking gate similar 
to that employed in radar tracking systems.^ 
The analog output of the tracker was propor- 
tional to distance between the crystals (trans- 
mission time) and was precalibrated in vitro. 
After low pass filtering the diameter signal was 
differentiated with a passive differentiator to 
obtain instantaneous rate of change of diame- 
ter. Maximum delay through both tracker and 
filter was approximately 10 msec and was con- 
stant up to 20 cps. End-diastolic diameter (Da) 
was measured at the beginning of isovolumic 
systole, and end-systolic diameter (Dg) was 
taken as the minimum diameter achieved dur- 
ing systole. Percent systolic shortening of inter- 
nal diameter was calculated as Da — Dg/Da . 
The rate of diameter change (dD/dt) was 
measured directly from the differentiated diam- 
eter signal ; maximum systolic — dD/ dt, which 
occurred at or shortly following peak aortic 
flow, was divided by the instantaneous diameter 
to obtain peak circumferential fiber shortening 
rate, Vcf, in circs/sec ( — irdD/dt) /ttD. 
Within- and between-groups means were 
compared by the appropriate Student's T-test, 
assigning a two-tailed significance level to the P 
value. 
At necropsy there was no evidence of intra- 
thoracic infection in animals included in the 
study. Adhesions between the lungs and heart 
were routinely present, and the intrathoracic 
portions of the cables were encased in fibrous 
tissue. The intraventricular transducers were 
covered with small fibrinous thrombi in the ma- 
jority of animals, but the presence of such 
thrombi did not correlate with transducer mal- 
function. 
RESULTS 
Control State 
Under control conditions, with dogs stand- 
ing on the treadmill, there were no significant 
differences between normal and denervated ani- 
mals in stroke volume, cardiac output, devel- 
oped LV pressure, peak LV dp/dt, maximum 
rate of systolic diameter shortening ( — dD/dt) , 
or VcF (Table I) . Mild tachycardia was present 
in normal animals (average 128 beats/min.) as 
well as in denervated subjects (average 140 
beats/min. P > .05). Average end-diastolic di- 
ameter was greater in the denervated animals 
(28 vs. 25.8 mm), but this difference was not 
statistically significant. There was, however, a 
significant difference between denervated and 
normal dogs in percent shortening of transverse 
diameter during systole, with normal animals 
